Nooo this is cruel and unusual punishment 😭😭😭

Thanks to to Noah for sticking it out with me through many years of geospatial travails, and to many others for help and feedback along the way.

We think that this conception of a constantly shifting topography of state capacity is potentially useful for a range of questions about the state and political violence.

We find that these negative shocks to state repressive capacity increased the likelihood of insurgent-state clashes. But this effect did not hold when local communities harbored strongly anti-insurgent attitudes, suggesting that state capacity and civilian behavior jointly shape rebel strategy.

We assess this idea in the context of the Colombian civil war. To identify exogenous variation in the repressive capacity of the Colombian state, we study road closures induced by natural landslides, which temporarily hindered the delivery of support and reinforcements to local security forces.

We propose that this consensus, if anything, understates the influence of state capacity. State repressive capacity varies not just between countries or across regions but in a fine-grained, constantly shifting topography, generating frequent windows of weakness insurgents can seize upon.

Thrilled that you're joining us Anton!

I love xG, but also I hate xG (you're telling me it could be bad luck OR we're terrible at finishing OR the keeper is playing out of their mind??)

Ah thank you Ana Paula!! Someday we're going to ask you to teach us how to make it Bayesian 😀

An early version of this paper started life a million years ago as my MA thesis at UChicago CIR, but was left behind until Sebastian stepped in to save it. I'm grateful to him for a terrific transatlantic co-authorship and to the many who provided us comments over the years.